System and method for testing high temperature of composite ceramic material intermediate frequency furnace

Abstract

The invention relates to the technical field of composite ceramic material performance testing, in particular to a system and method for testing the high temperature of a composite ceramic material intermediate frequency furnace, and the system comprises an intermediate frequency induction furnace main body, a vacuum control system, a water cooling system, a central control unit and a multi-source temperature sensing system. The multi-source temperature sensing system comprises an embedded high-temperature-resistant contact type temperature measuring unit. By constructing a multi-source temperature sensing system and adopting a self-adaptive data fusion algorithm, the long-term stability of embedded contact temperature measurement, the local interference resistance of two-color infrared temperature measurement and the full-field sensing capability of thermal imaging are creatively fused; on-line calibration reference is provided for the optical sensor by means of high precision of the embedded thermocouple in a medium-low temperature section, drift and lag of the thermocouple at extremely high temperature are compensated by means of the advantages of the optical sensor in a high temperature section, and system errors and random noise of a single sensor are systematically eliminated by means of a finally output fusion temperature value. And the overall temperature measurement precision is higher.

Description

Technical Field

[0001] This invention relates to the field of composite ceramic material performance testing technology, specifically to a testing system and method for high-temperature testing of composite ceramic materials in a medium-frequency furnace. Background Technology

[0002] Composite ceramic materials, with their excellent high-temperature resistance, high strength, corrosion resistance, and low thermal conductivity, have become core materials in strategic fields such as aerospace, nuclear energy, and high-end equipment manufacturing. The high-temperature performance of these materials directly determines their service safety and lifespan; therefore, it is necessary to use specialized testing systems to simulate extreme high-temperature environments and accurately evaluate their performance.

[0003] Medium-frequency induction heating furnaces have become the mainstream equipment for high-temperature performance testing of composite ceramic materials due to their fast heating rate, high thermal efficiency, and ease of achieving vacuum / inert atmosphere protection. Currently, traditional medium-frequency furnace high-temperature testing systems mainly consist of the furnace body, a single temperature measuring component (such as a tungsten-rhenium thermocouple or a monochromatic infrared thermometer), a simple atmosphere control unit, and a basic data recording module. However, some shortcomings remain in their application.

[0004] Reliability issues of contact temperature sensing elements at extreme high temperatures: To accurately measure temperature, high-temperature resistant contact sensors such as tungsten-rhenium (W / Re) thermocouples are usually used. However, at extreme high temperatures above 1800℃, especially when the test needs to be carried out in a specific active or oxidizing atmosphere, the thermocouple wire will undergo a severe oxidation reaction. The oxide layer formed on the surface causes the thermoelectric potential characteristics to drift, resulting in inaccurate temperature readings. At the same time, high temperatures can cause the thermocouple wire to recrystallize and become brittle, and the mechanical strength will drop sharply. It is very easy to break during furnace vibration or sample loading and unloading. This not only increases the cost of expensive consumables, but also causes test interruptions due to frequent sensor replacements, which seriously affects test efficiency and continuity.

[0005] Non-contact infrared thermometry is hampered by the uncertainty of material emissivity: While infrared thermometry avoids contact issues, its principle relies on the key parameter of the emissivity of the surface of the object being measured. Composite ceramic materials are typically composed of multiple phases, and their surface emissivity is not a fixed constant but changes dynamically and significantly with increasing temperature, surface oxidation, phase transformation, sintering densification, and other processes. Existing infrared thermometers usually require a fixed emissivity value to be preset, which cannot adapt to this dynamic change in real time. This results in systematic errors in the temperature measurement results that are difficult to calibrate, failing to meet the needs of high-precision material testing.

[0006] The inherent temperature non-uniformity problem of induction heating: Medium-frequency induction heating relies on the eddy current effect generated by electromagnetic induction. Its "skin effect" causes heat to be concentrated on the surface of the sample. For composite ceramic materials with low thermal conductivity, the huge internal and external temperature difference can induce severe thermal stress, and even cause the sample to crack before testing. In addition, the electromagnetic field distribution generated by traditional induction coils has inherent non-uniformity, which will further aggravate the radial and axial temperature gradients of the sample. This makes the test environment deviate greatly from the ideal state, and the obtained material performance data cannot truly reflect the material properties. Instead, it includes the error introduced by the heating process, resulting in poor reproducibility and comparability of the test.

[0007] Therefore, a testing system and testing methods for high-temperature induction furnaces of composite ceramic materials are needed to improve the above-mentioned problems. Summary of the Invention

[0008] The purpose of this invention is to provide a testing system and method for high-temperature testing of composite ceramic materials in medium-frequency furnaces, in order to solve the problems mentioned in the background art.

[0009] To achieve the above objectives, the present invention provides the following technical solution:

[0010] A testing system for high-temperature applications in medium-frequency furnaces for composite ceramic materials includes a medium-frequency induction furnace body, a vacuum control system, a water cooling system, a central control unit, and a multi-source temperature sensing system. The multi-source temperature sensing system includes:

[0011] An embedded high-temperature contact temperature measuring unit includes at least one tungsten-rhenium thermocouple encapsulated in a high-purity ceramic sleeve. The ceramic sleeve extends and is installed through a channel pre-embedded in the refractory material of the furnace sidewall, with its temperature measuring end positioned adjacent to the sample bearing area.

[0012] A dual-color infrared thermometer unit includes at least one dual-color infrared thermometer, wherein the probe of the dual-color infrared thermometer is positioned facing the sample surface through a first observation window protected by a first purge air curtain.

[0013] Through the above technical solutions, the two-color method can, to a certain extent, reduce the effects of surface emissivity changes and partial shading.

[0014] A high dynamic range infrared thermal imaging unit includes at least one infrared thermal imager, the probe of which is positioned to face the entire sample area through a second observation window protected by a second purge air curtain.

[0015] The data fusion and processing core is electrically connected to the multi-source temperature sensing system and is configured to synchronously collect data from each sensor and perform temperature field reconstruction and multi-source data fusion calculation.

[0016] As a preferred embodiment of the present invention, the ceramic sleeve in the embedded high-temperature contact temperature measuring unit is made of yttrium-stabilized zirconia, and the combination of the ceramic sleeve and the thermocouple is a modular and detachable structure.

[0017] In the above technical solution, the sleeve is installed by extending through a channel pre-embedded in the refractory material of the furnace side wall, and its temperature measuring end is set near the sample bearing area to provide a stable and reliable contact temperature reference.

[0018] As a preferred embodiment of the present invention, the airflow provided by the first and second Purge air curtains is an inert gas compatible with the furnace atmosphere, used to keep the viewing window lens clean and to disperse volatiles on the optical path.

[0019] In the above technical solution, the gas supply device for the first and second purging air curtains includes an inert gas storage tank, a mass flow controller, and an air curtain nozzle. The inert gas is argon or nitrogen with a purity ≥99.999%. The flow rate of the first purging air curtain is controlled at 5-8 L / min, and the flow rate of the second purging air curtain is controlled at 8-12 L / min. The air curtain nozzle has a ring structure and is sleeved on the outside of the observation window. The nozzle outlet forms a 30° angle with the observation window lens, forming an oblique airflow along the surface of the lens. Both the first and second observation windows are made of sapphire material with a thickness of 5-8 mm and a light transmittance ≥90% (wavelength 1-5 μm). A detachable quartz protective sheet is provided on the outside of the lens.

[0020] As a preferred embodiment of the present invention, the core of data fusion and processing includes:

[0021] The temperature field reconstruction module is configured to generate and display a three-dimensional temperature field distribution map of the sample area based on the two-dimensional thermal image data acquired by the high dynamic range infrared thermal imaging unit and through a three-dimensional reconstruction algorithm.

[0022] The multi-source data fusion and adaptive calibration module is configured to run an adaptive weighted fusion algorithm to dynamically integrate the output data of the embedded temperature measurement unit, the dual-color infrared temperature measurement unit, and the temperature field reconstruction module to obtain and output an optimal fused temperature value.

[0023] The temperature field reconstruction module is configured as follows: based on the 2D thermal image data acquired by the high dynamic range infrared thermal imaging unit, combined with the three-dimensional model of the sample, the data of the edge region of the thermal image is completed by interpolation algorithm, and then a 3D temperature field distribution map of the sample region is generated by three-dimensional mapping algorithm. The temperature field resolution is consistent with that of the thermal imager and supports real-time rendering on the display interface of the central control unit.

[0024] The multi-source data fusion and adaptive calibration module is configured to: run an adaptive weighted fusion algorithm, receive in real time the temperature value T1 of the embedded temperature measurement unit, the temperature value T2 of the dual-color infrared temperature measurement unit, and the average temperature value T3 of the sample center region output by the temperature field reconstruction module, calculate the real-time uncertainty of each data, and dynamically allocate weights w1, w2, and w3, where w1 + w2 + w3 = 1, and finally fuse the temperature value T_fusion = w1T1 + w2T2 + w3T3.

[0025] As a preferred embodiment of the present invention, the multi-source data fusion and adaptive calibration module adopts an adaptive weighted Kalman filter algorithm and dynamically adjusts the weight of each sensor data in the fusion calculation based on the real-time uncertainty estimate of each sensor data.

[0026] The adaptive weighted Kalman filter algorithm in the above technical solution specifically includes:

[0027] Prediction phase: Based on the previous fusion temperature T_fusion(k-1) and the rate of change of heating power of the intermediate frequency furnace, predict the current temperature T_predict(k);

[0028] Update phase: Calculate the residuals between the measured and predicted values ​​of each sensor: e1 = T1 - Tpre(k), e2 = T2 - Tpre(k), e3 = T3 - Tpre(k), based on the residual variance σ1. 2 =var(e1), σ2 2 =var(e2), σ3 2 =var(e3) assigns weights w i =1 / σ i 2 / (1 / σ1 2 +1 / σ2 2 +1 / σ3 2 );

[0029] Output phase: The current fusion temperature Tfusion(k) = w1T1 + w2T2 + w3T3 is used to update the Kalman gain matrix.

[0030] As a preferred embodiment of the present invention, the multi-source data fusion and adaptive calibration module is further configured to: use the fusion temperature value to reverse calculate the real-time emissivity of the sample surface, and feed back the calculated emissivity value to the high dynamic range infrared thermal imaging unit to correct the measurement results in its monochromatic temperature measurement mode.

[0031] In the above technical solution, the multi-source data fusion and adaptive calibration module is further configured as follows:

[0032] Emissivity Inversion: Based on Planck's Blackbody Radiation Formula I = εσT 4 σ = 5.67 × 10 -8 W / (m 2 ·K4 Using the sample surface radiation intensity I (corresponding to the thermal image grayscale value, which has been calibrated) acquired by the thermal imaging unit and the fusion temperature Tfusion, the real-time emissivity ε = I / (σTfusion) is calculated. 4 The inversion accuracy is ≤ ±0.02; feedback correction: ε is fed back to the infrared thermal imaging unit via RS485 interface at a frequency of 1Hz. The thermal imager automatically updates the emissivity parameters of the internal temperature measurement algorithm, recalculates the temperature of each pixel, and updates the 3D temperature field distribution map.

[0033] A test method for a high-temperature testing system for intermediate frequency furnaces of composite ceramic materials, the method comprising the following steps:

[0034] S1. No-load calibration procedure:

[0035] With no sample in the furnace, start the vacuum system to evacuate to the target vacuum level (or introduce the target atmosphere at a flow rate of 100-200 sccm), and then start the water cooling system.

[0036] The medium-frequency furnace heats up to the calibrated temperature points at a rate of 5-10℃ / min, including 500℃, 1000℃, 1500℃ and 2000℃, and holds at each temperature point for 10-15 minutes;

[0037] Using the measurement value of the embedded contact temperature measurement unit as the reference T, the deviation ΔTred = Tred - Treference and the deviation ΔTheat = Theat - Treference of the dual-color infrared temperature measurement value Tred from Treference are calculated, and the deviation ΔTheat = Theat - Treference of the thermal imager center temperature measurement value Theat from Treference are written into the parameter configuration module of the corresponding unit as calibration coefficients.

[0038] S2. Online Measurement and Fusion Steps:

[0039] Place the composite ceramic sample, where the sample size is ≤ 80% of the stage area. Start the vacuum system and water cooling system, and heat according to the set temperature curve, where the heating rate is 5~50℃ / min, the holding temperature is 800~2000℃, and the holding time is 10~120min.

[0040] The multi-source temperature measurement system synchronously acquires data: T1 (0.5s / time), T2 (0.1s / time), and thermal imaging data (0.2s / frame);

[0041] The data fusion core performs 3D temperature field reconstruction, runs the fusion algorithm to output T_fusion, and displays the T_fusion and temperature field distribution map in the central control unit.

[0042] S3. Cooling and Data Export Steps:

[0043] After the heat preservation is completed, the temperature is reduced at a rate of 5-20℃ / min. When the temperature drops below 500℃, the medium frequency furnace is turned off and water cooling is continued until the temperature reaches room temperature.

[0044] Export test data (temperature-time curve, 3D temperature field snapshot, fused temperature value) for material property analysis.

[0045] As a preferred solution of the present invention, in the online measurement and fusion step, the data fusion algorithm mainly relies on the data of the embedded contact temperature measurement unit in the low temperature stage, and dynamically increases the weight of the reconstructed data of the two-color infrared temperature measurement unit and the high dynamic range infrared thermal imaging unit in the high temperature stage.

[0046] In the above technical solution, in the low temperature stage (T fusion < 1000 °C): the weight w1 of the embedded contact temperature measurement unit is 0.6 - 0.7, the two-color infrared w2 is 0.2 - 0.3, and the weight w3 of the thermal image reconstruction T3 is 0.1 - 0.2;

[0047] In the high temperature stage (T fusion ≥ 1000 °C): w1 gradually decreases to 0.2 - 0.3 to avoid high temperature oxidation drift of the thermocouple, w2 = 0.3 - 0.4, w3 = 0.3 - 0.4. If the temperature uniformity of the monitored sample by the thermal imager ≤ ±3 °C, w3 is increased to 0.5 and w2 is decreased to 0.2. If the uniformity > ±3 °C, w2 is increased to 0.5 and w3 is decreased to 0.2.

[0048] As a preferred solution of the present invention, the method includes:

[0049] Emissivity feedback step: Use the fused temperature value to inversely calculate the real-time emissivity of the sample surface and use it to correct the measurement result of the high dynamic range infrared thermal imaging unit.

[0050] In the above technical solution, step S2 further includes an emissivity feedback step S2a:

[0051] The data fusion core calculates the real-time emissivity ε = I / (σT fusion^4) every 1 s and feeds it back to the infrared thermal imager through RS485;

[0052] The thermal imager immediately updates the emissivity parameter and recalculates the pixel point temperature. For example, after the oxidation of SiC-based ceramic at 1500 °C, ε increases from 0.58 to 0.72, and the temperature correction of the thermal imager changes from 1420 °C to 1500 °C;

[0053] The corrected thermal image data is used to update the 3D temperature field distribution map to ensure the accuracy of the temperature field.

[0054] As a preferred solution of the present invention, the method includes: uniformity calculation step: Based on the three-dimensional temperature field distribution map, calculate and output the temperature uniformity statistical index of the sample area in real time.

[0055] In the above technical solution, step S2 further includes a uniformity calculation step S2b:

[0056] Based on the 3D temperature field distribution map, the temperature of all pixels in the sample area is extracted, and the following indicators are calculated:

[0057] Temperature difference ΔT = Tmax - Tmin;

[0058] Temperature standard deviation (n is the number of pixels);

[0059] 90% confidence interval temperature range, of which the temperature range after removing the 5% highest temperature and 5% lowest temperature;

[0060] The uniformity index is output every 10 seconds, generating a "time-uniformity" curve to evaluate the rationality of the heating process.

[0061] Compared with the prior art, the beneficial effects of the present invention are:

[0062] 1. In this invention, a multi-source temperature sensing system is constructed and an adaptive data fusion algorithm is adopted to creatively integrate the long-term stability of embedded contact temperature measurement, the anti-local interference of dual-color infrared temperature measurement, and the full-field perception capability of thermal imaging. The high precision of embedded thermocouples in the medium- and low-temperature range provides an online calibration reference for the optical sensor. Furthermore, the advantages of the optical sensor in the high-temperature range are utilized to compensate for the drift and hysteresis of the thermocouple at extreme temperatures. The final fused temperature value systematically eliminates the systematic errors and random noise of a single sensor, resulting in higher overall temperature measurement accuracy.

[0063] 2. In this invention, through a real-time emissivity feedback correction mechanism, the true emissivity of the sample surface can be calculated in reverse based on the blackbody radiation law using a high-precision fusion temperature value, and dynamically fed back to the infrared thermal imager at a frequency of 1Hz. This allows the thermal imager to update its internal algorithm parameters in real time. Even if the emissivity of the material surface changes from 0.58 to 0.72 due to oxidation, sintering, or other processes, the system can automatically correct it, ensuring the accuracy of single-point and full-field temperature measurement results. This reduces the temperature measurement error caused by emissivity uncertainty from more than ±5% in the traditional method to within ±1%.

[0064] 3. In this invention, the data fusion core based on adaptive weighted Kalman filtering can intelligently evaluate the real-time reliability of each sensor's data and dynamically allocate weights. When a sensor malfunctions, the system can automatically reduce its weight to ensure the robustness of the output results. At the same time, the modular embedded thermocouple design and the efficient Purge air curtain protection system greatly extend the service life and maintenance cycle of the sensors, reduce test interruptions, and lower operating costs. Attached Figure Description

[0065] Figure 1 This is a schematic diagram of the system startup and no-load calibration process of the present invention;

[0066] Figure 2 This is a schematic diagram of the core data fusion algorithm of the present invention;

[0067] Figure 3 This is a schematic diagram of the Purge air curtain and optical measurement protection process of the present invention; Detailed Implementation

[0068] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0069] To facilitate understanding of the present invention, a more comprehensive description of the invention will be given below with reference to the accompanying drawings, and several embodiments of the invention will be provided. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that the disclosure of the invention will be more thorough and complete.

[0070] For examples, please refer to Figure 1-3 The present invention provides a technical solution:

[0071] A testing system for high-temperature applications in medium-frequency furnaces for composite ceramic materials includes a medium-frequency induction furnace body, a vacuum control system, a water cooling system, a central control unit, and a multi-source temperature sensing system. The vacuum control system, water cooling system, and central control unit are commonly used structures in medium-frequency induction furnaces and are well-known techniques in this field, so they will not be elaborated upon here. The multi-source temperature sensing system includes:

[0072] An embedded high-temperature contact temperature measuring unit includes at least one tungsten-rhenium thermocouple encapsulated in a high-purity ceramic sleeve. The ceramic sleeve extends and is installed through a channel pre-embedded in the refractory material of the furnace sidewall, with its temperature measuring end positioned adjacent to the sample bearing area.

[0073] A dual-color infrared thermometer unit includes at least one dual-color infrared thermometer, wherein the probe of the dual-color infrared thermometer is positioned facing the sample surface through a first observation window protected by a first purge air curtain.

[0074] In the above technical solution, the main body of the medium-frequency furnace includes a furnace shell, a furnace lining, an induction coil, and a sample stage. The furnace lining is made of zirconium oxide-alumina composite refractory material with a thickness of 80-120mm, which can withstand temperatures above 2200℃. The induction coil is wound around the outside of the furnace lining and is made of copper tubing with an inner diameter of 8-12mm and an outer diameter of 12-16mm. The induction coil adopts a spiral staggered winding structure, with an axial stagger distance of 5-10mm between adjacent turns and a circumferential stagger angle of 5-10°. The number of turns of the induction coil is 10-15. The staggered winding structure can weaken the electromagnetic field gradient and reduce the temperature gradient caused by uneven electromagnetic field. The sample stage is located in the center of the furnace lining and is made of silicon carbide. A rotary drive mechanism is connected to the bottom of the sample stage, which can drive the sample stage to rotate at a speed of 5-10r / min, so that all parts of the sample are heated evenly.

[0075] Through the above technical solutions, the two-color method can, to a certain extent, reduce the effects of surface emissivity changes and partial shading.

[0076] A high dynamic range infrared thermal imaging unit includes at least one infrared thermal imager, the probe of which is positioned to face the entire sample area through a second observation window protected by a second purge air curtain.

[0077] The data fusion and processing core is electrically connected to the multi-source temperature sensing system and is configured to synchronously collect data from each sensor and perform temperature field reconstruction and multi-source data fusion calculation.

[0078] In this embodiment, the ceramic sleeve in the embedded high-temperature contact temperature measuring unit is made of yttrium-stabilized zirconia, and the combination of the ceramic sleeve and the thermocouple is a modular and detachable structure.

[0079] In the above technical solution, the sleeve is installed by extending through a channel pre-embedded in the refractory material of the furnace side wall, and its temperature measuring end is set near the sample bearing area to provide a stable and reliable contact temperature reference.

[0080] In this embodiment, the airflow provided by the first and second Purge air curtains is an inert gas compatible with the furnace atmosphere, used to keep the viewing window lens clean and disperse volatiles on the optical path.

[0081] In the above technical solution, the gas supply device for the first and second purging air curtains includes an inert gas storage tank, a mass flow controller, and an air curtain nozzle. The inert gas is argon or nitrogen with a purity ≥99.999%. The flow rate of the first purging air curtain is controlled at 5-8 L / min, and the flow rate of the second purging air curtain is controlled at 8-12 L / min. The air curtain nozzle has a ring structure and is sleeved on the outside of the observation window. The nozzle outlet forms a 30° angle with the observation window lens, forming an oblique airflow along the surface of the lens. Both the first and second observation windows are made of sapphire material with a thickness of 5-8 mm and a light transmittance ≥90% (wavelength 1-5 μm). A detachable quartz protective sheet is provided on the outside of the lens.

[0082] High-purity inert gas prevents impurities from being introduced into the observation window by the gas curtain, and is compatible with the inert / reducing atmosphere inside the furnace (such as hydrogen-argon mixture) to prevent the gas curtain from reacting with the gas inside the furnace to generate pollutants (such as H2O, CO2).

[0083] Differentiated flow design: The first air curtain (dual-color infrared) has a low flow rate to avoid disturbing the local temperature field on the sample surface, while the second air curtain (thermal imaging) has a high flow rate to cover a larger optical path and effectively disperse volatiles;

[0084] The spiral airflow generated by the 30° angled nozzle flows along the surface of the lens, reducing the impact on the furnace temperature field compared to a direct-fired air curtain. The sapphire viewing window is resistant to high temperatures, and the quartz protective sheet reduces maintenance costs.

[0085] In this embodiment, the data fusion and processing core includes:

[0086] The temperature field reconstruction module is configured to generate and display a three-dimensional temperature field distribution map of the sample area based on the two-dimensional thermal image data acquired by the high dynamic range infrared thermal imaging unit and through a three-dimensional reconstruction algorithm.

[0087] The multi-source data fusion and adaptive calibration module is configured to run an adaptive weighted fusion algorithm to dynamically integrate the output data of the embedded temperature measurement unit, the dual-color infrared temperature measurement unit, and the temperature field reconstruction module to obtain and output an optimal fused temperature value.

[0088] The temperature field reconstruction module is configured as follows: based on the 2D thermal image data acquired by the high dynamic range infrared thermal imaging unit, combined with the three-dimensional model of the sample, the data of the edge region of the thermal image is completed by interpolation algorithm, and then a 3D temperature field distribution map of the sample region is generated by three-dimensional mapping algorithm. The temperature field resolution is consistent with that of the thermal imager and supports real-time rendering on the display interface of the central control unit.

[0089] The multi-source data fusion and adaptive calibration module is configured to: run an adaptive weighted fusion algorithm, receive in real time the temperature value T1 of the embedded temperature measurement unit, the temperature value T2 of the dual-color infrared temperature measurement unit, and the average temperature value T3 of the sample center region output by the temperature field reconstruction module, calculate the real-time uncertainty of each data, and dynamically allocate weights w1, w2, and w3, where w1 + w2 + w3 = 1, and finally fuse the temperature value T_fusion = w1T1 + w2T2 + w3T3.

[0090] Among them, the computing power of the embedded processor meets the real-time requirements of 3D temperature field reconstruction (latency ≤100ms), avoiding the lag in temperature field display caused by insufficient computing power of traditional industrial computers (reduced from 500ms to 100ms), and ensuring that testers can grasp the dynamic temperature distribution of the sample in a timely manner.

[0091] Compared with linear interpolation, Kriging interpolation improves the accuracy of thermal image edge data completion by 30%, and can fully reflect the temperature of sample edge areas (such as the corners of sheet-like samples), avoiding temperature field distortion caused by field distortion.

[0092] The dynamic weighting mechanism adjusts the contribution based on sensor reliability (e.g., w1 drops from 0.5 to 0.2 when T1 drifts), which improves temperature measurement accuracy by 25% compared to fixed weight fusion (fusion temperature error ≤ ±1℃).

[0093] In this embodiment, the multi-source data fusion and adaptive calibration module adopts an adaptive weighted Kalman filter algorithm and dynamically adjusts the weight of each sensor data in the fusion calculation based on the real-time uncertainty estimate of each sensor data.

[0094] The adaptive weighted Kalman filter algorithm in the above technical solution specifically includes:

[0095] Prediction phase: Based on the previous fusion temperature T_fusion(k-1) and the rate of change of heating power of the intermediate frequency furnace, predict the current temperature T_predict(k);

[0096] Update phase: Calculate the residuals between the measured and predicted values ​​of each sensor: e1 = T1 - Tpre(k), e2 = T2 - Tpre(k), e3 = T3 - Tpre(k), based on the residual variance σ1. 2 =var(e1), σ2 2 =var(e2), σ3 2 =var(e3) assigns weights w i =1 / σ i 2 / (1 / σ1 2 +1 / σ2 2 +1 / σ3 2 );

[0097] Output phase: The current fusion temperature Tfusion(k) = w1T1 + w2T2 + w3T3 is used to update the Kalman gain matrix.

[0098] Among them, the "prediction-update" closed-loop characteristic of Kalman filtering effectively suppresses random noise from sensors (such as fluctuations caused by airflow disturbances in infrared thermometry), improving noise suppression capability by 50% and reducing the fusion temperature fluctuation range from ±3℃ to ±1℃.

[0099] Weighting based on residual variance can identify sensor anomalies in real time, such as σ² caused by contamination of the dual-color infrared observation window. 2 Increase w2 and automatically decrease w2 to avoid abnormal data dominating the fusion results, improving the system's anti-interference ability by 40%.

[0100] The algorithm has a simple logic, is compatible with the computing power of embedded processors, requires no additional hardware, and reduces system costs.

[0101] In this embodiment, the multi-source data fusion and adaptive calibration module is further configured to: use the fusion temperature value to calculate the real-time emissivity of the sample surface in reverse, and feed back the calculated emissivity value to the high dynamic range infrared thermal imaging unit to correct the measurement results in its monochromatic temperature measurement mode.

[0102] In the above technical solution, the multi-source data fusion and adaptive calibration module is further configured as follows:

[0103] Emissivity Inversion: Based on Planck's Blackbody Radiation Formula I = εσT 4 σ = 5.67 × 10 -8 W / (m 2 ·K 4 Using the sample surface radiation intensity I (corresponding to the thermal image grayscale value, which has been calibrated) acquired by the thermal imaging unit and the fusion temperature Tfusion, the real-time emissivity ε = I / (σTfusion) is calculated. 4 The inversion accuracy is ≤ ±0.02.

[0104] Feedback correction: ε is fed back to the infrared thermal imaging unit via RS485 interface at a frequency of 1Hz. The thermal imager automatically updates the emissivity parameters of the internal temperature measurement algorithm, recalculates the temperature of each pixel, and updates the 3D temperature field distribution map.

[0105] Among them, real-time emissivity inversion solves the infrared temperature measurement error caused by the dynamic change of emissivity of composite ceramics at high temperatures (such as the increase of ε from 0.58 to 0.72 after oxidation of SiC-based ceramics), and improves the monochromatic temperature measurement accuracy from ±5% to ±1%.

[0106] No additional dedicated monochrome infrared thermometer is required; the thermal imager combines "temperature field distribution" and "high-precision single-point temperature measurement", reducing system costs by 30%.

[0107] The 1Hz feedback cycle responds promptly to changes in the sample surface condition (such as oxide layer formation and sintering shrinkage), avoiding cumulative errors caused by emissivity lag (error is reduced from 80℃ to ±5℃), and improving temperature measurement accuracy by 90% throughout the entire testing cycle.

[0108] A test method for a high-temperature testing system for intermediate frequency furnaces of composite ceramic materials, the method comprising the following steps:

[0109] S1. No-load calibration procedure:

[0110] With no sample in the furnace, start the vacuum system to evacuate to the target vacuum level (or introduce the target atmosphere at a flow rate of 100-200 sccm), and then start the water cooling system.

[0111] The medium-frequency furnace heats up to the calibrated temperature points at a rate of 5-10℃ / min, including 500℃, 1000℃, 1500℃ and 2000℃, and holds at each temperature point for 10-15 minutes;

[0112] Using the measurement value of the embedded contact temperature measurement unit as the reference T, the deviation ΔTred = Tred - Treference and the deviation ΔTheat = Theat - Treference of the dual-color infrared temperature measurement value Tred from Treference are calculated, and the deviation ΔTheat = Theat - Treference of the thermal imager center temperature measurement value Theat from Treference are written into the parameter configuration module of the corresponding unit as calibration coefficients.

[0113] S2. Online Measurement and Fusion Steps:

[0114] Place the composite ceramic sample, where the sample size is ≤ 80% of the stage area. Start the vacuum system and water cooling system, and heat according to the set temperature curve, where the heating rate is 5~50℃ / min, the holding temperature is 800~2000℃, and the holding time is 10~120min.

[0115] The multi-source temperature measurement system synchronously acquires data: T1 (0.5s / time), T2 (0.1s / time), and thermal imaging data (0.2s / frame);

[0116] The data fusion core performs 3D temperature field reconstruction, runs the fusion algorithm to output T_fusion, and displays the T_fusion and temperature field distribution map in the central control unit.

[0117] S3. Cooling and Data Export Steps:

[0118] After the heat preservation is completed, the temperature is reduced at a rate of 5-20℃ / min. When the temperature drops below 500℃, the medium frequency furnace is turned off and water cooling is continued until the temperature reaches room temperature.

[0119] Export test data (temperature-time curves, 3D temperature field snapshots, fusion temperature values) for material performance analysis.

[0120] In this embodiment, in the online measurement and fusion step, the data fusion algorithm mainly relies on the data of the embedded contact temperature measurement unit at low temperature stages, and dynamically increases the weights of the reconstructed data of the two-color infrared temperature measurement unit and the high-dynamic-range infrared thermal imaging unit at high temperature stages.

[0121] In the above technical solution, at low temperature stages (T_fusion < 1000 °C): the weight w1 of the embedded contact temperature measurement unit is 0.6 - 0.7, the two-color infrared w2 is 0.2 - 0.3, and the weight w3 of the thermal image reconstruction T3 is 0.1 - 0.2;

[0122] At high temperature stages (T_fusion ≥ 1000 °C): w1 gradually drops to 0.2 - 0.3 to avoid high-temperature oxidation drift of the thermocouple, w2 = 0.3 - 0.4, w3 = 0.3 - 0.4. If the temperature uniformity of the sample monitored by the thermal imager is ≤ ±3 °C, w3 is increased to 0.5 and w2 is decreased to 0.2. If the uniformity > ±3 °C, w2 is increased to 0.5 and w3 is decreased to 0.2.

[0123] Among them, the different weights in the high and low temperature stages are adapted to the accuracy characteristics of each sensor: relying on contact temperature measurement at low temperature and infrared sensors at high temperature, the fusion accuracy in the full temperature range is ≤ ±1.5 °C;

[0124] Adjust w2 and w3 in combination with the temperature uniformity to ensure the temperature accuracy in key test scenarios. Compared with fixed weights, the fusion error in the high temperature stage is further reduced by 15%.

[0125] In this embodiment, the method includes:

[0126] Emissivity feedback step: Using the fused temperature value to inversely calculate the real-time emissivity of the sample surface, and used to correct the measurement results of the high-dynamic-range infrared thermal imaging unit.

[0127] In the above technical solution, step S2 further includes an emissivity feedback step S2a:

[0128] The data fusion core calculates the real-time emissivity ε = I / (σT_fusion^4) every 1 s and feeds it back to the infrared thermal imager through RS485;

[0129] The thermal imager immediately updates the emissivity parameter and recalculates the pixel point temperature. For example, after the oxidation of SiC-based ceramics at 1500 °C, ε rises from 0.58 to 0.72, and the temperature correction of the thermal imager changes from the original 1420 °C to 1500 °C;

[0130] The corrected thermal image data is used to update the 3D temperature field distribution map to ensure the temperature field accuracy.

[0131] Among them, the temperature measurement deviation of the thermal imager is corrected in real time to avoid temperature underestimation / overestimation caused by changes in emissivity;

[0132] The corrected 3D temperature field more accurately reflects the temperature distribution of the sample, providing reliable data support for analyzing the thermal stress distribution of materials.

[0133] In this embodiment, the method includes: a uniformity calculation step: based on the three-dimensional temperature field distribution map, calculating and outputting the temperature uniformity statistical index of the sample area in real time.

[0134] In the above technical solution, step S2 further includes a uniformity calculation step S2b:

[0135] Based on the 3D temperature field distribution map, the temperature of all pixels in the sample area is extracted, and the following indicators are calculated:

[0136] Temperature difference ΔT = Tmax - Tmin;

[0137] Temperature standard deviation (n is the number of pixels);

[0138] 90% confidence interval temperature range, of which the temperature range after removing the 5% highest temperature and 5% lowest temperature;

[0139] The uniformity index is output every 10 seconds, generating a "time-uniformity" curve to evaluate the rationality of the heating process.

[0140] Among them, the multi-dimensional uniformity index comprehensively reflects the temperature distribution of the sample: ΔT intuitively reflects the maximum temperature difference, σT reflects the degree of dispersion, and the 90% confidence interval excludes extreme interference points. Compared with calculating only ΔT, the scientific nature of the evaluation is improved by 50%.

[0141] Real-time uniformity monitoring can optimize heating parameters in a timely manner and reduce material property testing errors caused by temperature unevenness;

[0142] The "time-uniformity" curve provides a basis for subsequent optimization of the testing process and improves test repeatability.

[0143] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A testing system for high-temperature applications in medium-frequency furnaces for composite ceramic materials, comprising a medium-frequency induction furnace body, a vacuum control system, a water cooling system, and a central control unit, characterized in that, Also includes: A multi-source temperature sensing system, the system comprising: An embedded high-temperature resistant contact temperature measuring unit includes at least one tungsten-rhenium thermocouple encapsulated in a high-purity ceramic sleeve. The ceramic sleeve extends and is installed through a channel pre-embedded in the refractory material of the furnace sidewall, and its temperature measuring end is located adjacent to the sample bearing area. A dual-color infrared thermometer unit, comprising at least one dual-color infrared thermometer, wherein the probe of the dual-color infrared thermometer is positioned facing the sample surface through a first observation window protected by a first purge air curtain. A high dynamic range infrared thermal imaging unit, comprising at least one infrared thermal imager, wherein the probe of the infrared thermal imager is positioned to face the entire sample area through a second observation window protected by a second purge air curtain. The data fusion and processing core is electrically connected to the multi-source temperature sensing system and is configured to synchronously collect data from each sensor and perform temperature field reconstruction and multi-source data fusion calculation.

2. The testing system for high-temperature testing in a medium-frequency furnace for composite ceramic materials according to claim 1, characterized in that, The ceramic sleeve in the embedded high-temperature contact temperature measuring unit is made of yttrium-stabilized zirconia, and the combination of the ceramic sleeve and the thermocouple is a modular and detachable structure.

3. The testing system for high-temperature testing in a medium-frequency furnace for composite ceramic materials according to claim 1, characterized in that, The first and second purgatory air curtains provide airflows of inert gas compatible with the furnace atmosphere, used to keep the viewing window lens clean and disperse volatiles on the optical path.

4. The testing system for high-temperature testing in a medium-frequency furnace for composite ceramic materials according to claim 1, characterized in that, The core of the data fusion and processing includes: The temperature field reconstruction module is configured to generate and display a three-dimensional temperature field distribution map of the sample area based on the two-dimensional thermal image data acquired by the high dynamic range infrared thermal imaging unit and a three-dimensional reconstruction algorithm. The multi-source data fusion and adaptive calibration module is configured to run an adaptive weighted fusion algorithm to dynamically integrate the output data of the embedded temperature measurement unit, the dual-color infrared temperature measurement unit, and the temperature field reconstruction module to obtain and output an optimal fused temperature value.

5. The testing system for high-temperature testing in a medium-frequency furnace for composite ceramic materials according to claim 4, characterized in that, The multi-source data fusion and adaptive calibration module adopts an adaptive weighted Kalman filter algorithm and dynamically adjusts the weight of each sensor data in the fusion calculation based on the real-time uncertainty estimate of each sensor data.

6. The testing system for high-temperature testing in a medium-frequency furnace for composite ceramic materials according to claim 4 or 5, characterized in that, The multi-source data fusion and adaptive calibration module is further configured to: use the fused temperature value to calculate the real-time emissivity of the sample surface in reverse, and feed back the calculated emissivity value to the high dynamic range infrared thermal imaging unit to correct the measurement results in its monochromatic temperature measurement mode.

7. A test method using the test system for high-temperature testing of composite ceramic materials in a medium-frequency furnace as described in any one of claims 1-6, characterized in that, The method includes the following steps: No-load calibration steps: With no sample in the furnace, heat to multiple calibration temperature points, and automatically calibrate the measurement parameters of the dual-color infrared temperature measurement unit and the high dynamic range infrared thermal imaging unit based on the measurement values ​​of the embedded high-temperature resistant contact temperature measurement unit. Online measurement and fusion steps: During sample testing, data from the multi-source temperature sensing system are collected synchronously, a three-dimensional temperature field is reconstructed, and a data fusion algorithm is run to output high-precision fused temperature values ​​and temperature field distribution information.

8. The method according to claim 7, characterized in that, In the online measurement and fusion step, the data fusion algorithm mainly relies on the data from the embedded contact temperature measurement unit in the low-temperature stage, and dynamically increases the weight of the reconstructed data from the dual-color infrared temperature measurement unit and the high dynamic range infrared thermal imaging unit in the high-temperature stage.

9. The test method according to claim 7, characterized in that, The method includes: Emissivity feedback step: The real-time emissivity of the sample surface is calculated by inverting the fusion temperature value and used to correct the measurement results of the high dynamic range infrared thermal imaging unit.

10. The test method according to claim 7, characterized in that, The method includes: a uniformity calculation step: based on the three-dimensional temperature field distribution map, calculating and outputting the temperature uniformity statistical index of the sample area in real time.

Images